Multi-site cross-organ calibrated deep learning (MuSClD): Automated diagnosis of non-melanoma skin cancer.

Details

Ressource 1Download: 36516556_Multi-site cross-organ calibrated deep learning.pdf (18010.99 [Ko])
State: Public
Version: Final published version
License: CC BY-NC-ND 4.0
Serval ID
serval:BIB_576D871C7998
Type
Article: article from journal or magazin.
Collection
Publications
Institution
Title
Multi-site cross-organ calibrated deep learning (MuSClD): Automated diagnosis of non-melanoma skin cancer.
Journal
Medical image analysis
Author(s)
Zhou Y., Koyuncu C., Lu C., Grobholz R., Katz I., Madabhushi A., Janowczyk A.
ISSN
1361-8423 (Electronic)
ISSN-L
1361-8415
Publication state
Published
Issued date
02/2023
Peer-reviewed
Oui
Volume
84
Pages
102702
Language
english
Notes
Publication types: Journal Article ; Research Support, N.I.H., Extramural ; Research Support, Non-U.S. Gov't ; Research Support, U.S. Gov't, Non-P.H.S.
Publication Status: ppublish
Abstract
Although deep learning (DL) has demonstrated impressive diagnostic performance for a variety of computational pathology tasks, this performance often markedly deteriorates on whole slide images (WSI) generated at external test sites. This phenomenon is due in part to domain shift, wherein differences in test-site pre-analytical variables (e.g., slide scanner, staining procedure) result in WSI with notably different visual presentations compared to training data. To ameliorate pre-analytic variances, approaches such as CycleGAN can be used to calibrate visual properties of images between sites, with the intent of improving DL classifier generalizability. In this work, we present a new approach termed Multi-Site Cross-Organ Calibration based Deep Learning (MuSClD) that employs WSIs of an off-target organ for calibration created at the same site as the on-target organ, based off the assumption that cross-organ slides are subjected to a common set of pre-analytical sources of variance. We demonstrate that by using an off-target organ from the test site to calibrate training data, the domain shift between training and testing data can be mitigated. Importantly, this strategy uniquely guards against potential data leakage introduced during calibration, wherein information only available in the testing data is imparted on the training data. We evaluate MuSClD in the context of the automated diagnosis of non-melanoma skin cancer (NMSC). Specifically, we evaluated MuSClD for identifying and distinguishing (a) basal cell carcinoma (BCC), (b) in-situ squamous cell carcinomas (SCC-In Situ), and (c) invasive squamous cell carcinomas (SCC-Invasive), using an Australian (training, n = 85) and a Swiss (held-out testing, n = 352) cohort. Our experiments reveal that MuSCID reduces the Wasserstein distances between sites in terms of color, contrast, and brightness metrics, without imparting noticeable artifacts to training data. The NMSC-subtyping performance is statistically improved as a result of MuSCID in terms of one-vs. rest AUC: BCC (0.92 vs 0.87, p = 0.01), SCC-In Situ (0.87 vs 0.73, p = 0.15) and SCC-Invasive (0.92 vs 0.82, p = 1e-5). Compared to baseline NMSC-subtyping with no calibration, the internal validation results of MuSClD (BCC (0.98), SCC-In Situ (0.92), and SCC-Invasive (0.97)) suggest that while domain shift indeed degrades classification performance, our on-target calibration using off-target tissue can safely compensate for pre-analytical variabilities, while improving the robustness of the model.
Keywords
Humans, Deep Learning, Australia, Skin Neoplasms/diagnostic imaging, Skin Neoplasms/pathology, Carcinoma, Basal Cell/pathology, Carcinoma, Squamous Cell/diagnostic imaging, Carcinoma, Squamous Cell/pathology, Deep learning, Digital pathology, Domain adaptation, Domain shift, Non-melanoma skin cancer, Subtyping
Pubmed
Web of science
Open Access
Yes
Create date
19/12/2022 10:58
Last modification date
19/12/2023 8:20
Usage data